Analysis of Forest Stand Resistance to Insect Attack According to Remote Sensing Data
Round 1
Reviewer 1 Report
Forests 1322665
Analysis of forest stand resistance to insect attack according to remote sensing data
This manuscript uses remote sensing data from fir stands attacked by two insect pests to assess stand susceptibility to environmental change. Due to the remote nature of large tiaga forests, remote sensing via satellite data is useful. Standardizing methods is necessary to make such a tool useful across multiple different forests and landscapes. While the idea is useful, the manuscript is poorly written and needs further development. Aspects of the results are not explained in a way that the reader can follow.
Abstract
L 10 – Language is awkward at times. First line of the abstract would read better if it was “We propose a method for analysing the state of forest plantations in Siberian tiaga forests using remote sensing data.”
L 11 – define “state” – do the authors mean health status? Growth state?
L 11 – Like above, language is awkward. Better as “As an indicator of state, normalized difference vegetative index (NDVI) and radiation temperature of the territory (LST) were used to formulate a proposed index of sensitivity. The index was…”
L 15 – “…two fir test sites…”
L 17, 19 – Scientific name authorities (Tschetv and Fischer) should not be italicized. Also, Tschetv is an abbreviation and needs a period.
Introduction
L 27-34 – The introduction should set the justification based on the available literature. The first paragraph contains no cited literature – Statements should be supported by literature.
L 27 – Over use of “the” – it is rarely necessary in English. This line can easily be “Control and monitoring of forest stand states in boreal zones is extremely difficult due to size and inaccessibility of forest areas.” Same goes for the next sentence – “Insect outbreaks are one of the main factors in forest stand weakening.” There are times when it is needed for the sentence (like L 35).
L 35 – “territories” is unnecessary
L 53-57 – There should be a clear statement of the objectives of the study or hypotheses to be tested.
Materials and Methods
L 59 – Just state what you did – no need to couch it in obscuring language. “We used quantitative measurements of plant biomass based on…”
L 64 – “To date” is unnecessary
Fig 1 – check y-axis labels
Fig 1 and 2 seem like they could be combined to side-by-side figure.
L 164 – Extra break
Figure 5 missing years of data collection (before and after need years).
L 198 – Indentions missing for several upcoming paragraphs. Also, bad form to use a question here. Just make the statements based on the data analysis in the results (the questions being asked should be what makes up the objectives and hypotheses as the end of the introduction).
Figure 6 should have an indication of
L 210 – Results exist after the study has been completed – meaning they all need to be past tense.
L 233, 238 – Bad form. Make the statement regarding shapes, patterns, etc. in the figure, then cite it.
L 240 – “…significant deviations of the response function spectrum in future 240 outbreaks from control stands were observed.” – I don’t see it. The data points are all different year scales, so it is not clear to the reader where this deviation is in response in Figure 7. Same with Figure 8.
Figure 7 and 8 – Don’t assume the reader knows that the black triangle is the outbreak year – state it in the caption for the figure. Since the points represent different time frames, this is not easy to interpret. The author needs to explain the figure in far more detail. Also, reference to the figure should come before the figure.
Discussion
This is the most poorly developed discussion I have read reviewing a manuscript. The authors did nothing to interpret their proposed index in the context of the literature.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Dear authors
The discussion is a bit short and the conclusions are vague (you have to guess at them), so I propose to expand it by trying to answer the following questions:
Does the forecast concern the possibility of primary harmful leaf and secondary stem insects (which can be determined with a given probability) in the coming year or several years ahead?
Can you also predict how long the insect outbreak will last?
What is the uncertainty of the analysis? Does it only apply to thermophilic insects? Can it replace the currently used autumn search for pests in forest litter or the cutting of trees into sheets to determine insect species, the number of females or, for example, the degree to which pupae are parasitised by fungi, etc.?
Does the data have to be acquired from satellites or can multispectral cameras be installed on drones or planes? Is it possible to identify tree species in mixed stands and is the prediction then equally good?
Can chlorophyll fluorescence measurements, as they are inversely proportional to photosynthetic capacity, also be used?
Is it possible to identify weakened trees or groups of trees which are most vulnerable to insect attack, for example, in order to remove them in breeding operations?
What are the strengths and weaknesses of the proposed method?
Can it replace the existing methods or rather be a complement to them?
Author Response
"Please see the attachment."
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Forests 1322665 v2
Analysis of forest stand resistance to insect attack according to remote sensing data
I still think this is a good idea (using remote sensing to assess susceptibility of inaccessible tiaga forests). The authors have addressed many of my original comments in an appropriate and acceptable manner. However, I do have suggestions for revisions below. Some items were in my original review and not addressed.
One issue I still have is the lacking definition of what “state” the authors are referring to – what forest stand state?
Another issue is the lack of defined objectives or hypothesis tested at the end of the introduction.
Finally, the discussion still provides no context. The discussion in a scientific paper is for interpreting the results in the context of the existing knowledge – that knowledge exists because of published literature. There is no way the authors can interpret the usefulness of applying such a remote sensing technique without citing the existing literature within the field. How does this align with Safonova et al. (2019, Remote Sensing doi:10.3390/rs11060643)? This is a paper discussing the use of UAV to assess fir trees attacked by bark beetles – remote sensing (UAV instead of satellites), fir, Russia, etc. However, the authors of this paper do not cite Safonova et al. to interpret their results in the context of the existing literature. How about Rees et al. (2002, Ambio Special Report 12, 56-62) to set the context in the introduction of how difficult it is to monitor even boundary changes in the tiaga? What about Kharuk et al. (2007, Scandinavian Journal of Forest Research 22, 531-536) discussing spatial patterns of Siberian silkmoth (one of the pests the authors cover in this manuscript!)? The authors do not use this as a way to interpret their results or place it in context of the existing literature.
I’m not asking for just the insertion of citations – I’m suggesting that the discussion needs to be written in a way that set context and interprets the results within that context based on the literature.
L 29 – I still think there is a need to define “states.” Health status? Since they talk about management in L 35, is this stocking, production, or biomass state?
L 37 – Again, what is the stand state of interest?
L 38 – [From original review] “territories” is unnecessary
L 94-97 – [From original review] There should be a clear statement of the objectives of the study or hypotheses to be tested. This is still missing.
Fig 1 – There is still an error in the y-axis label for NDVI.
L 161 – Bad form to say NDVI and LST are shown in Figs. 1 (note that it is no longer 2 figures). Make a statement about the figure, then cite it. “NDVI and LST follow similar patterns, increasing in values until reaching a peak at 200 day of the year and then decreasing in values (Fig. 1).”
Author Response
Dear Reviewer. We have taken into account your comments as much as possible.
Abstract: The non-obvious term "condition" is replaced by "resistance of stand to insect attacks". Changes in the transfer function between time series input (LST) and output time series (NDVI) make a significant contribution to stand resistance. These changes are shown statistically significant before insect attacks (in the results).
Introduction: added a discussion of the works (Safonova et al., 2019) и (Kharuk et al., 2007) and their difference from the proposed method.
Discussion: added a discussion of the reasons causing change in the spectral response function.
Fig.1 – As mentioned above in the text "NDVI - dimensionless indicator in range (0-1)." Perhaps that's enough?
Best regards, authors.
